The overall focus of the proposed studies in this application is to determine how cytotoxic function following allogeneic bone marrow transplantation influences the outcome of the transplant. Experiments will address related questions utilizing cytotoxically defective donor strains of mice. Single (i.e., perforin or CD95L = csd) and double cytotoxic deficient (cdd) strains will be employed to assess the role of donor cytotoxic function in GVH responses contributing to progenitor cell engraftment and GVH disease using experimental BMT models involving MHC mismatched and minor histocompatibility antigen mismatch donor/recipient combinations. Studies are designed to generate mouse strains deficient in tumor necrosis factor receptor-1 and/or 2 expression to be used to establish an allogeneic BMT model lacking three major effector pathways mediating cytotoxicity. These models will be used to analyze GVHD induced weight loss, clinical signs, tissue damage and immune deficiency. The use of CD4+ and CD8+ T cell subsets in this model will be of interest to determine if the "subtraction" of these effector pathways differentially affects their capacity to induce GVH responses. The "triple deficient" model can subsequently be used for the design of experiments to address the potential contribution of additional cytotoxic pathways involving TWEAK and TRAIL dependent signaling. These approaches will permit further testing of the hypothesis that differing effector molecules play more and less important roles in differing GVHD target compartments. Finally, experiments are proposed to examine the potential application of infused recipient (i.e., syngeneic) populations to kill donor cells following BMT to inhibit or reverse ongoing GVHD. Cell populations to be examined will include cytotoxic T cells and antigen presenting cells transduced with death receptor ligands.